How to Use CTR to Measure Funnel Conversion
- CTR = (Clicks / Impressions) x 100. It measures the decision to click and nothing that happens after it.
- Click rate, click-through rate and click-to-open rate use three different denominators, so the same email campaign produces three different numbers.
- Conversion rate is the product of the step rates before it: a 7 percent CTA click-through rate and a 15 percent form completion rate give a 1.05 percent page conversion rate.
- A CTR that rises while conversion rate stays flat is a message-match failure between the element and the page behind it, not a win.
- There is no universal good CTR. Placement and intent decide the number, so the only useful benchmark is the element's own baseline.
Click-through rate is clicks divided by impressions. It answers one question: of the people who saw this element, how many chose to click it? Conversion rate answers a different question: of the people who could have completed the goal, how many did? Neither number replaces the other, and a funnel measured with only one of them is measured badly.
Click-through rate was once mostly an advertising metric, used to grade paid campaigns. On-site it does a different and more useful job. It shows which design and copy elements earn attention, where in the funnel visitors stop moving forward, and whether a behavior pattern is real before you commit an experiment to the revenue metric. This article covers the formula with a worked example, the exact difference between click-through rate, click rate and conversion rate, where to measure each stage, and a case study where a click goal was the right goal.
What click-through rate is, and how to calculate it
Worked example
A landing page is viewed 18,000 times in a month. Its primary call to action, a button that opens a download form, is clicked 1,260 times.
CTR = 1,260 ÷ 18,000 × 100 = 7%
Of those 1,260 clickers, 189 complete the form. The step conversion rate after the click is 189 ÷ 1,260 = 15%. The conversion rate of the page itself is 189 ÷ 18,000 = 1.05%.
Notice what the third number is made of. 7% × 15% = 1.05%. The page conversion rate is the product of the step rates in front of it, which is why a single conversion rate can move for reasons that have nothing to do with the page you are looking at. Break it into its steps and each movement has an address.
Two rules keep the calculation honest. Use unique clicks against unique impressions, so one enthusiastic visitor clicking six times does not inflate the numerator. And count an impression only when the element was genuinely on screen, not merely present in the page code. An element far below the fold that most visitors never scroll to will show a flattering CTR against a denominator it never earned.
Click-through rate vs click rate vs conversion rate
Most of the confusion around these terms is a denominator problem, not a definition problem. The numerator is nearly always "clicks". What changes is what you divide by, and that choice decides what the number means.
| Metric | Formula | What it tells you | Where it is used |
|---|---|---|---|
| Click-through rate (CTR) | Unique clicks ÷ impressions | How persuasive an element is to the people who actually saw it | Ads, search results, on-site CTAs, banners, filters |
| Click rate | Unique clicks ÷ messages delivered | How much of the whole list acted, including people who never opened | Email and SMS reporting |
| Click-to-open rate (CTOR) | Unique clicks ÷ unique opens | How well the content worked on the people it reached | Email content and offer testing |
| Conversion rate | Goal completions ÷ visitors or sessions | Whether the outcome you are paid for actually happened | Landing pages, checkout, whole-funnel reporting |
Two practical consequences follow. First, never compare a click rate from one tool with a click-through rate from another and call it a trend. Second, some platforms report a total click rate that counts repeat clicks by the same person, which is why a "click rate" can occasionally exceed the unique rate for the same send. Check the definition in your own reporting before you compare anything.
If you want the metric definitions on their own, see click-through rate and conversion rate in our glossary.
What CTR tells you that conversion rate does not
A single metric can never cover the full range of a marketer's goals. Conversion rate is the outcome, and outcomes are the point, but an outcome is the sum of many small decisions and it hides all of them. When the conversion rate of a landing page falls by a fifth, that number alone cannot tell you whether fewer people saw the call to action, fewer clicked it, or fewer completed the form behind it. The three causes have three different fixes.
CTR splits that outcome into decisions. Combined with conversion rate across the funnel's stages, it gives you reasons rather than only numbers. Specifically, CTR does three things well:
- It reveals behavioral habits. Which route visitors take through the site, which entry points they favor, whether they browse a category tree or go straight to search.
- It localizes the drop-off. A stage with healthy traffic and a poor CTR is where interest is lost, and it is where the next test belongs.
- It measures engagement cheaply. Clicks accumulate far faster than purchases, so a click goal reaches a usable sample size in a fraction of the time a revenue goal needs.
Where to measure CTR in the conversion funnel
The top of an eCommerce funnel contains strangers attracted by SEO, paid advertising, social media and email. They become visitors when they land on a page and start acting on it. Everything after that is a sequence of clicks, and each click has a rate you can measure.
-
Entry: the ad, listing or search resultImpressions are what the platform served; clicks are the visits you paid for or earned. A low CTR here is usually a promise problem in the title and description, not a landing page problem.
-
Landing: the hero call to actionDenominator is the visitors who reached the page. This is the single most useful on-site CTR to track, because it separates "nobody arrives" from "everybody arrives and nobody engages".
-
Navigation: category links, site search and filtersSearch and filter usage tells you what visitors expected to find and could not. Segment traffic by whether people used search, and the two groups usually behave very differently.
-
Consideration: product page actionsAdd to cart, size guide, reviews, delivery information. Each has its own CTR, and a high CTR on a reassurance element is a signal about an unanswered objection.
-
Decision: cart and checkout buttonsThe CTR of the proceed-to-checkout button against cart views isolates hesitation at the moment of commitment, before payment friction enters the picture.
-
Return: email and remarketing linksReport click rate and click-to-open rate side by side. One grades the list and the send, the other grades the content.
For the wider view of how these stages fit together, see our guide to conversion funnel analysis.
Set a click as the primary goal and the purchase as the secondary goal in the same test.
Learn more about Omniconvert Explore →The call to action is what CTR actually measures
What the CTA does for the visitor
- It answers "where do I click for this?" A visually distinct button removes the search. Design changes alone can move CTR, and they are straightforward to validate with A/B testing software.
- It answers "what is in it for me?" That is the button copy's job, and copy usually moves CTR more than styling does.
What the CTA does for the marketer
- It moves visitors from one funnel stage to the next, which is what makes its CTR a stage metric.
- It reveals which design and copy elements appeal to which visitors.
- It shows what the audience expects from the site's offers, wording and layout.
- It produces a clean, fast-accumulating goal for experiments.
The distinction to hold on to: conversion rate counts the visitors who completed a task, such as subscribing, downloading or buying. CTR counts the clicks on a specific element. Measure the CTA's conversion rate alone and you cannot tell whether the call to action or the page content is doing the work. Measure both and you can. There is more on the wording, sizing and placement of buttons in our guide to CTA optimization.
Case study: using a click goal to validate a behavior pattern
The analysis started in Google Analytics with a simple segmentation of traffic into two groups, which the team called "searchers" and "non-searchers". Comparing the conversion rates of the two segments showed that visitors who used site search converted at a much higher rate than those who did not. Site-search users often convert far better than other visitors, so check this split in your own analytics before you build on it. The client's absolute conversion figures cannot be disclosed.
A correlation like that has two possible explanations, and they lead to opposite decisions. Either people who search are already high-intent, in which case searching is a symptom, or searching itself helps people find what they want, in which case encouraging it is a lever. The only way to tell is to change the behavior and watch what happens.
The experiment was set up like this:
- Hypothesis: changing the design of the search bar will increase the number of searches by 10%.
- Traffic segment: all visitors.
- Primary goal: number of searches, defined as a pageview goal counting visitors who used the search bar. This is a click-through rate on the search CTA.
- Change tested: the control search bar against a variation with an added green button prompting visitors to search.
Searches on the variation rose by 15%, comfortably above the 10% in the hypothesis, and the conversion rate difference between the control and the variation was 3%.
One warning belongs with this result. The test was deliberately aimed at the number of searches rather than at revenue, because the client had other live experiments running and the team did not want to compromise their accuracy. That is a legitimate reason to choose a click goal, but it also means this test validated a behavior, not a revenue effect. Testing the impact of different site search versions on the site's conversion rate is a separate experiment.
Which metric you measure is a prioritization decision tied to your own objective. Sometimes the objective is to raise conversion rate now. Sometimes, as here, it is to confirm that a pattern is real so that later experiments start from solid ground. Visitors follow patterns on the way to converting. Your job is to identify the pattern, then test it rather than assume it.
What CTR and conversion rate look like together in real experiments
| Brand | What was tested | Effect on clicks | Effect on the outcome |
|---|---|---|---|
| University of London | CTA button, copy and position on the course page | +15.68% click-through, at 99% statistical relevance | +45.26% applications |
| Bonia | Removing competing homepage CTAs in favor of one prioritized path | +200% click-through | +218% conversion rate on women's watches, at over 97% statistical relevance |
| ING Bank Romania | CTA and application form rewritten around a fear surfaced by user research | +60% click-through rate | +20% leads |
| eCommerce client (site search) | Adding a search button to the search bar | +15% searches | 3% difference in conversion rate between control and variation |
Read the last two columns against each other every time. When they agree, you have a repeatable win. When the click column moves and the outcome column does not, the element is attracting clicks that the page behind it cannot convert, and the fix is on that page, not on the button.
Which visitors clicked matters as much as how many. A first-time visitor and a repeat customer respond to different wording, so a blended CTR can hide two opposite results. Segmenting customers by value and behavior in Nexus by Omniconvert gives you the audiences to read those results apart.
When CTR is the wrong metric
Three situations where a click goal misleads:
- Paid campaigns judged on cost per acquisition. A higher CTR that brings in unqualified traffic raises spend and lowers return. The click is a cost, not a result.
- Curiosity copy. A vague or exaggerated label reliably wins the click test and loses on the next page. If CTR is up and bounce rate on the destination is up with it, that is what happened.
- Elements with tiny denominators. A footer link seen by few visitors can post an impressive percentage on a handful of clicks. Always read the CTR next to the raw impression count.
The useful summary is the same one the original version of this article ended on, and it still holds: measuring CTR is not useful when the conversion matters more than the click. Measuring CTR is useful to validate behavior patterns and measure engagement.
Frequently Asked Questions
Click-through rate is the share of people who saw something clickable and clicked it. It is calculated as clicks divided by impressions, multiplied by 100. The clickable thing can be an ad, a search result, an email link, a call-to-action button, a banner, a filter or a search bar. CTR measures one step only: the decision to click. It says nothing about what the visitor does after the click.
CTR = (Clicks / Impressions) × 100. If a landing page is viewed 18,000 times in a month and its primary call to action is clicked 1,260 times, the CTR is 1,260 divided by 18,000, which is 7 percent. Use unique clicks and unique impressions on both sides of the formula, and count the impression only when the element was actually visible, not just present in the page code.
Click-through rate measures how many people clicked an element out of everyone who saw it. Conversion rate measures how many people completed the goal out of everyone who had the chance to complete it. CTR is a step metric and conversion rate is an outcome metric. They multiply together: if 7 percent of visitors click the call to action and 15 percent of those clickers finish the form, the page converts at 1.05 percent.
Not always, and the difference matters in email reporting. Click-through rate divides clicks by impressions, meaning opportunities to click. Click rate in most email platforms divides unique clicks by messages delivered, so people who never opened the email are still in the denominator. Click-to-open rate divides unique clicks by unique opens instead. Three different denominators produce three different numbers from the same campaign, so always confirm which one your tool reports before you compare.
There is no universal good CTR, because the number depends entirely on placement, intent and audience. A branded search result, a checkout button and a display banner produce CTRs that are orders of magnitude apart, and none of them is comparable to the others. The only benchmark that means anything is your own: measure the current CTR of the specific element, then test changes against that baseline.
Because a click is easy to buy and a conversion is not. Curiosity copy, an unclear label or an overstated promise all raise CTR by attracting people who were never going to convert, and the mismatch shows up as a bounce on the next page. When CTR rises and conversion rate stays flat or falls, the element is winning the click and losing the promise. That is a message-match problem between the element and the page behind it.
Measure CTR at every point where a visitor has to choose to move forward: the ad or search result that brings them in, the hero call to action, category and product links, the site search bar and filters, the add-to-cart button, the checkout button, and the links inside your emails. Each of those is a funnel stage with its own denominator, and a stage-level CTR shows you where attention stops, which a single site-wide conversion rate cannot.
Yes, and it is often the right goal to test when the click itself is the behavior you want to validate. Set the click as the primary goal in your testing tool, then track the downstream conversion rate as a secondary metric so you can see whether the extra clicks turned into extra outcomes. Omniconvert Explore records both in the same experiment, which is what stops a CTR win from hiding a conversion loss.
Pick the one element that carries the most traffic in your funnel: the hero call to action, the add-to-cart button, or the site search bar. Measure its click-through rate for the last full month, then measure the conversion rate of the people who clicked it. Those two numbers together tell you whether your problem is attention or persuasion, and they point at completely different fixes. If few people click, work on visibility, wording and placement. If many people click and few convert, work on the page behind the click. Then run one A/B test with the click as the primary goal and the conversion as the secondary goal, so that whichever way the result lands, you learn something you can act on.
Test what earns the click, and what happens after it
Omniconvert Explore runs A/B and multivariate tests with click goals and conversion goals in the same experiment, so you can validate a behavior pattern without losing sight of revenue. It also runs on-site surveys, so you can ask the visitors who did not click why they did not.